Ludwig Phase II Clone Debugging and Miscellany

[digi2t]

Hey R.G., what does your sim say if we take R18 (RF18 in the clone) down to, say, 5K. It's a 470K now.

I've been fiddling again, but this time with the downstream stages. My R4 idea works, but I found that the fuzz was a little anemic. Probably due to the lower input signal hitting Q2. Same story when I fiddled with the resistor to ground on Q3. But, on Q4, taking the base down further started to reduce the distortion in the clean FFM mode. As for the fuzz, it's actually starting to reflect the original. The clone, as is, has got great sustain. Much more than the original. It actually surprised me when I started playing with it, how meaty the fuzz was. I've now taken R18 down to 5K, and the clean FFM signal cleaned up, and the fuzz is thinner now, more like the original. Maybe Q4 distorting was driving both over the top, helping to accentuate the fuzz more? Also, it doesn't affect the Voice Fuzz, or Fuzz Repeats modes as much. When playing with the other resistor values to ground on Q2, and Q3, I would either start to lose in one, the other, or again, on the fuzz. I think I might be onto a happy middle ground here.

I compared the voltages on Q4 as well with the 5K, and they're still pretty close to the original.

As soon as I get the big volume tests done, I'll report what works best.

[R.G.]

Hey R.G., what does your sim say if we take R18 (RF18 in the clone) down to, say, 5K. It's a 470K now.

No need to look at the sim. Changing that should have a huge effect on the DC bias of QF4. If we're talking about the same Rf18 on Qf4 in the clone, the bias is set by Rf17 and Rf18 as a voltage divider to about 1/3 of the voltage on Qf4 collector. Changing Rf18 to 5K would be dramatic if I understand the schemo right.

Can you check that the voltages don't change? That's a bit bizarre.

What happens if you use 4.7K in series with 22uF parallel with the existing 470K? This would knock the AC down a lot but not affect the DC conditions.

[digi2t]

OK, R26 is the one to replace with a trimmer, or lower value resistor. All the others either weaken the fuzz, weaken both the fuzz and the clean FFM signal too much, or weaken the fuzz repeat mode. Adjusting R26 can bring down just the clean FFM, allowing me to balance the signal between the clean bypass, clean FFM, and Fuzz.

If you bring the input (R6) closer to ground, you weaken everything, bypassed volume included. No good.

If you bring Q3 or Q4 base resistors closer to ground, it weakens both the FFM and the Fuzz. The fuzz really suffers, getting really thin with much less sustain, when dialing the FFM down to where it cleans up. Dialing down R18 hurts the Fuzz repeat mode, especially in Voice Fuzz, so that's out as well.

Replacing R26 with a small Bourns 3296 50K trimmer is the best workaround, without weakening the fuzz. Or, one could just use a 50K pot to find the best resistance, and then slip the appropriate resistor in there. This allows you to dial down the clean signal from hitting the filters too hard, and not affect the fuzz works. 

No need to mod the board either. Tie the wiper to one leg or the other, and there is plenty of room to place the trimmer in place of the resistor.

My two original units had the same problem. Super loud clean FFM, to the point of distortion if I don't turn down the guitar volume. I'm even debating mod'ing my original unit now. Maybe if I find some carbon comp resistors that match the look of the originals. 

   

[R.G.]

R26 it is then. This is 'rf26' in the clone, correct?

Through sheer good luck, that happens to be in a part of the PCB where there is actually room to stick in a trimmer footprint. If there's ever any more of these, I'll stick a trimmer footprint in there in parallel with the hard-resistor footprint so it can be trimmed as an option.

[digi2t]

R26 it is then. This is 'rf26' in the clone, correct?

Through sheer good luck, that happens to be in a part of the PCB where there is actually room to stick in a trimmer footprint. If there's ever any more of these, I'll stick a trimmer footprint in there in parallel with the hard-resistor footprint so it can be trimmed as an option.

Good luck?? PHHHWWTTT! I'm sure you did that on purpose... unwittingly  .

Yes, that's RF26. The extra parallel trimmer footprint would be the cherry.

Thanks R.G.

*EDIT* - This just dawned on me. I didn't think to bring it up before, but it might be good to mention as well that some folks might even want to add an external pot on the unit to control R26 resistance. Why? Because different guitar pickups will drive this signal differently. I'm using guitars that range between 7K single coils, all the way to 27.5K humbuckers. The hotter the pups, the more it will distort. Pretty sure they didn't have any pups that hot back in the day, so we're dealing with the engineering remnants here. If you're only using one guitar, then pick your resistor and run with it. If you've got a stable of them, then consider wiring an adjustment pot to the side of the case. Somewhere where it won't get knocked about, but you can access it for "guitar switch" adjustment.

Or maybe a 4 throw rotary. Hot, Hotter, Hottest, and Aye Caramba! settings.   

[digi2t]

OK, after many hours of pounding the keyboard, I think I've managed to cross reference all the trannies on the fall board in this puppy. The list below lists the transistors found on the fall board schematic, and their (I believe!)modern counterparts. Mind you, the clone works fine with what R.G. spec's in the build document. I just thought it would be cool to do a little detective work, and see what I could come up with. Here we go, Ludwig part number, followed by modern equivalent(s);

43054 - 2N5088 (I have these in the front end right now, and they work well. They also get my voltages much closer to the original.)
43045 - MPS8097 (hard to find), or BC546B (readily available, reverse pinout)
43175 - 2SC3332 (On semi still makes it, I managed to get my hands on 10.)
43176 - 2SA1319 (2SC3332's PNP counterpart. On semi still makes them, got my hands on 10.)
43173 - 2N5551 (Good call R.G.)
43174 - 2N5401 (See previous comment  )

Like I said, the spec'd trannies get the job done. I just wanted to test myself here, but there are some out there that might be curious (read "anal"  ) about what might be closer to the original. I've got the 2N5088's, 2N5551, and 2N5401's in mine now, with the others on their way. Once I get them, I'll plunk them in, and do some more sound/voltage comparisons with the original.

[Vince_b]

I'm about to wire the beast but I've got a question. How should I wire the power source? I'm not really familiar with: "Pwr Com", "AC/+DC1" and "AC/+DC2".

[digi2t]

I'm about to wire the beast but I've got a question. How should I wire the power source? I'm not really familiar with: "Pwr Com", "AC/+DC1" and "AC/+DC2".

Hi Vince,

If you are using the recommended 48 volt DC power adapter, then Pwr Com will be your negative, and connect the positive to either the +DC1 OR the +DC2. Do not connect the positive to both pads.

Looking forward to hearing your beast.

Cheers,
Dino

[Vince_b]

Thanks for the quick reply Dino!

As soon as I get it to work (or not) you will hear from me.

[Vince_b]

I'm not sure if it already have been mentionned, but the filter board is missing a hole for a wire. According to the wiring diagram, there should be a hole next to RF62 (trimmer) to put a wire that goes to the console board and to the rocker pot heel down. Is there something that I'm missing?

[R.G.]

Hmmmm. Lemme go check.

[R.G.]

@Vince:
having trouble locating that one. There is a circle for a wire on the diagram, but no hole there - which is, I think the inconsistency you noticed.

I believe this may be a non-problem. The wire location for that wire was over near RF62 on an earlier version, but moved. There is a quirk in the layout software that I use which forces me to manually make up the diagrams for wiring. I use the circles around wire pads to make that easier to locate, and since it's manual, sometimes I make errors in placing/removing a circle in doing the wiring artwork. It's a PITA, but cheaper than buying a $2000 'seat' for a fancier layout package.

@Dino:
You have more real experience with wiring that thing than I do. Have you encountered the issue?

[digi2t]

Yeah, I noticed that as well when I was wiring it. If you look at the "Parts Placement Diagram" page of the build guide, you'll notice the hole there, but there are no traces connecting it to anything. I assumed that things must have been moved around, and this was a remnant on the diagram, but deleted from production. On the wiring guide, simply delete the jumper shown between this hole and the other board (connecting to the "Rocker Pot Heel Down" wire).

That clear enough?

Cheers,
Dino

[R.G.]

Thanks. I located it. The leftover pad was from an early version of the board where I was still sorting pad and wire placement to make it easy (ier...) to wire controls. The traces to the pad got deleted, but the pad was left over. I'll delete it from the instructions and any future boards.

[Vince_b]

I have "finished" it, but I will have some debugging to do.
Every functions seem to work but the output volume is very low. I'm not sure yet about what is the cause.
The balance pot works only when the effect is bypassed, is it normal behavior? And the Lo-Z balance does nothing at all.
When I first plugged it, I reversed the polarity (I didn't noticed that the wall wart that I have bought was center positive), may it have destroyed something? I only let it run like that for less than a second though.
But that's enough for today, I will start proper debugging tomorrow.

[R.G.]

I have "finished" it, but I will have some debugging to do.
Every functions seem to work but the output volume is very low. I'm not sure yet about what is the cause.

With as many wires to connect as this thing has, a wiring mistake is one of the first places I'd look.

When I first plugged it, I reversed the polarity (I didn't noticed that the wall wart that I have bought was center positive), may it have destroyed something? I only let it run like that for less than a second though.

Probably not. There is a series diode with the power in to the board. That should have prevented damage. Taking transistor voltages should tell the tale.

[digi2t]

The balance pot works only when the effect is bypassed, is it normal behavior?

Yup, normal. This is just to balance the bypassed signal to the effected signal.

You may want to look over the first four transistor stages as well (Q1 - 4), for correct resistor values, and transistor voltages. Your problem might be here, if it`s not wiring.

The Lo-Z balance (if I remember correctly, but I might be wrong ) only works if you have two inputs going into the unit at the same time. This, I assume, allows you to achieve unity between the Hi and Lo signals.

Last time I looked at the build guide, I didn`t see it, but the input jacks should be mono switched jacks. With nothing plugged in, the input should be grounded. Otherwise, the unused input will generate noise.

[R.G.]

Build instructions updated, post later today.

[pinkjimiphoton]

nice work guys.

fwiw, dino's mod makes the unit *MUCH* easier to deal with. i think the fuzz may have been an afterthought to deal with how crappy the filter sounded without the fuzz on. i remember on my unit in the 70's, the bypass was very loud and treble boosted...almost like a %^&*ed wah at a point of oscillation.
dino's mod fixes and addresses this problem, and is highly reccomended...you can always dial in the original value if you WANT that sound.
but if you want it to play nice with modern standards, it's nite and day.

[Vince_b]

With as many wires to connect as this thing has, a wiring mistake is one of the first places I'd look.

That was my first thought too, but I've just double checked every wire connections and everything looks fine.

You may want to look over the first four transistor stages as well (Q1 - 4), for correct resistor values, and transistor voltages. Your problem might be here, if it`s not wiring.

I have checked the voltages of those transistors and according to the build document they seem right. The only thing that I have noticed is that when I put the switch on fast start, the emitter voltage of Q1 start to drop toward 0V. But I guess it is normal.

The fuzz is working, the filtering is working, the animation is working, the only problem is that I need to crank my amp just to get some almost decent volume. I will take a deeper look at it during this week-end.

And Dino, where did you get your pot for the expression pedal? I'm having a hard time finding one with enough threading to be able to secure it properly in the shell.